Resemble AI Launches Voice-Based Deepfake Simulation Platform

Resemble AI today launched a voice-based simulation platform to help organizations proactively test and strengthen their defenses against artificial intelligence-powered social engineering. Organizations using Resemble AI have reported a 90 percent reduction in attacks.

Resemble AI's platform uses hyper-realistic simulations to mirror how deepfake attacks unfold in the real world. The system runs voice phishing scenarios against employees, like a spoofed customer call asking for account access or an AI-generated voicemail claiming to be from a chief financial officer, and then tracks how they respond. Each employee is assigned a risk score from zero to 100, with team and organization level analytics to flag high-risk behaviors and blind spots.

Resemble AI's simulation platform is powered by its AI voice models and real-time large language model integration. Scenarios are delivered via phone calls, WhatsApp messages, and email, allowing security teams to simulate full attack chains across multiple channels.

Key features include the following:

Real-time voice cloning from just five seconds of audio.
Agentic AI that handles objections and maintains context.
Adaptive difficulty based on employee performance.
Policy-based AI risk scores per department that highlight where targeted training or safeguards are most needed.
Compliance reporting for internal security teams.

"Today's attackers aren't playing by the old rules. They're using sophisticated tactics and cloned voices to manipulate employees in real time," said Zohaib Ahmed, CEO and co-founder of Resemble AI, in a statement. "Our platform allows companies to test how those attacks would actually play out inside their walls and who's most at risk. It's been purposefully developed to be easy to learn through a training that teaches users to recognize deepfakes with a video and a quiz."